- cross-posted to:
- firefox@fedia.io
- technology@lemmy.world
- cross-posted to:
- firefox@fedia.io
- technology@lemmy.world
one passage of note:
Where does all of this leave the Firefox browser. Surman argued that the organization is very judicious about rolling AI into the browser — but he also believes that AI will become part of everything Mozilla does. “We want to implement AI in a way that’s trustworthy and benefits people,” he said. Fakespot is one example of this, but the overall vision is larger. “I think that’s what you’ll see from us, over the course of the next year, is how do you use the browser as the thing that represents you and how do you build AI into the browser that’s basically on your side as you move through the internet?” He noted that an Edge-like chatbot in a sidebar could be one way of doing this, but he seems to be thinking more in terms of an assistant that helps you summarize articles and maybe notify you proactively. “I think you’ll see the browser evolve. In our case, that’s to be more protective of you and more helpful to you. I think it’s more that you use the predictive and synthesizing capabilities of those tools to make it easier and safer to move through the internet.”
It’s kinda crazy to me how hate-filled Lemmy is just a half-year after I joined (when it felt like a breath of fresh air after deleting my Reddit account), especially for Mozilla. Mozilla has issues, but it’s nothing they can’t fix or come out of. They don’t deserve to die. And all the conspiracy theories I’m reading is just nonsense. I happen to be a Firefox user, but really it’s mainly because Google decided to screw with user choice (i.e. Manifest v3). Firefox is still FOSS, and it’s still giving plenty of user choice.
And all this AI talk is just bandwagoning by every corporation because if AI (as in LLM and whatnot) happens to be a baseline thing for many corporations, Mozilla not implementing it could backfire for them, so while it is bandwagoning, it also makes sense to hedge one’s bets on it.
I, for one, think this current notion of AI is too raw to take any real shape (outside of the current novelty), and these corps that are jumping on it, just like they did with “web 2.0” and “big data” and “the cloud” and “blockchain”, will eventually find that while there is some tangibility to be found, it will take many years to solidify into products that make sense for a consumer.
Yeah and AI is pretty useful for doing certain things. For example my pixel can turn on subtitles for any video or audio playing and even translate it for me on the fly. AI isnt blockchain and it isnt all chatgtp or making images with too many fingers. People are talking about improving web standards as if whatever ai stuff google,MS, and apple are cooking up wont be used in order to enhance various web features.
Likewise firefox is currently a good browser and does keep up for the most part. I’d understand the criticisms if firefox was suck in 2009, but modern day firefox is fast and works well and they will likely continue keeping up with standards while an independent team works on the open source AI stuff
Yep, exactly. Firefox does what it sets out to do – be a FOSS alternative to the ever-growing corporate browser market, providing user choice and useful features.
And I agree that AI as both a concept and implementation in its various types and forms will continue to be iterated upon and will continue to show numerous useful applications. I only hesitate that what is being called AI at this moment as being some kind of groundbreaking thing that changes the world, like corporations seem to be making it out as (hence the bandwagon to try to cash in on it).
It may very well change the world in the near future, but it’s not quite there yet. It is novel, and it is cool what ChatGPT and other applications can do, such as the example you gave, and that does have a very tangible quality to it. I feel that any fast-moving technological pace must be met with trust and with objective science every step of the way, and the likes of ChatGPT and other AI models have quite a ways to go – especially concerning trust.
I personally just don’t like people calling it AI. It’s not.
I happen to be a former
conkeror
user, which was based on XULRunner, which Mozilla dropped, just like other ways to use their engine for other browsers.But I could use pre-Australis Firefox.
Only then their experiments on people with UI design happened.
And yes, Mozilla as an organization has objective flaws. Their treatment of the SeaMonkey project (now separate) has not been good as well.
I realized after diving into this niche that depending on who you follow that’s exactly how people behave. I had to stop following @jwz because it’s only hate and negativity all the time. I have this exact same feeling on Mastodon.